As the title states. The program should be a simple comparison between the performance and accuracy of various data mining algorithms with and without encryption. I need the program to take the freely available newsgroups dataset, run it through encryption and classification and generate an excel which compares the accuracy and performance. You can choose the classification categories yourself but I need atleast 10 different categories.
1) The user interface will be extremely simple. Two dropdowns and a confirmation button.
a) First dropdown will have the list of the 4 encryption algos + an all option + none option
b) Second dropdown will have the list of the 4 classification algos + an all option
c) Press a button to run the selected combination.
2) The excel sheet should have the output of the above with one sheet per combination (for cases when user selects all) plus a sheet with the comparison of the output. The accuracy of the classification of all the combinations which were run + a comparison of the run times + a few charts showing the results.
I have a few algorithms in mind which i would like implemented but as speed is essential here if you can convince me to use algorithms of your choice that is also ok.
The project should be very easy to implement for someone who has experience with this type of work before. I will NOT give out the work to someone who cannot show me some example of relevant work that was done previously. Anyone who can show me a demo implementation with a working combination of one encryption and one classification algorithm which outputs the required data to an excel sheet gets higher priority automatically.
Also, code should be formatted and variables named properly. I will NOT accept code where there's no proper spacing or the variables are named with single letters and numbers.